The Effect of the Tax System as an Institutional Factor on the Business Structure in Europe
Bartha, Zoltan
University of Miskolc
19 June 2015
Online at https://mpra.ub.uni-muenchen.de/68027/
MPRA Paper No. 68027, posted 24 Nov 2015 05:33 UTC
Structure in Europe (chapter 7). In: A.S. Gubik & K. Wach (Eds.). Institutional Aspects of
Entrepreneurship. Miskolc: University of Miskolc, pp. 97-109.
7
The Effect of the Tax System as an Institutional Factor
on the Business Structure in Europe
Zoltan Bartha
University of Miskolc Faculty of Economics Institute of Economic Theory
Miskolc-Egyetemváros, H-3515 Miskolc, Hungary e-mail: zolib@hu.inter.net
Summary
This chapter focuses on the influence of tax systems and taxation rules on the firm structure of the 28 European Union member economies. It is argued that higher taxes and more complex tax rules lead to smaller firms, and that, on the other hand reduces macroeconomic performance. It is found that the firm size and corporate tax rates are negatively correlated in case of medium- sized (R=-0.4; p=0.03) and large (R=-0.41; p=0.03) firms. Indications were found that higher transaction costs caused by taxation lead to smaller firms, as a significant negative correlation was found between the number of hours per year needed to administer tax payments, and the share of large firms (R=-0.41 & p=0.03), and also between KPMG’s comment length (an indica- tor for tax system complexity) and the share of medium-sized firms’ turnover from the total turnover (R=-0.39 & p=0.045). More complicated tax rules might also cause a smaller proportion of firms to grow quicker, but the significance level of these relationships is not very convincing.
Keywords: tax system; tax rate; institutions; firm size JEL classifications: H21, L11
7.1. INTRODUCTION
Firm size is widely regarded as a key determinant of economic performance. Several studies have showed that larger firms generally outperform smaller ones: Pervan & Visic found a weak but significant positive relation between firm size and firm profitability (2012); Lee using panel data for more than 7,000 US publicly held firms found that profit
rates are positively correlated with firm size (2009). Larger firms also tend to be more internationalised (Gubik, 2007, 2008; Gubik & Bartha, 2014), and being involved in inter- national business processes usually also leads to better performance (Gubik, 2011). The relation between the innovativeness and the size of the firm is a lot more debatable (see Vaona & Pianta, 2006 for a fairly detailed review), but it is generally found that larger firms spend more on R&D, and can be more efficient in implementing innovations if certain conditions hold (Symeonidis, 1996). Smaller firms on the other hand can be more flexible, giving them an advantage over large enterprises (Urbaniec, 2015).
For the reasons mentioned above the firm structure of an economy is of great im- portance. The size of the firm can be influenced by a lot of factors: bigger economies offer a larger domestic market so firms can grow bigger; more labour, capital or other factors of production can also boost the size of the firm; traditional links to international markets (like the ones formed between some European countries and their former do- minions) can also lead to a larger firm size.
Institutional factors, the rules of doing business formed by the traditions and the po- litical bodies of a country, yet again can influence firm size. If it is very complicated to establish a new enterprise, already existing firms can grow bigger because of the lack of competition they have to face. Rules that make the domestic market monopolistic affect many areas of the economy, but they undoubtedly have an effect on the size of the do- mestic firms as well. Some rules disproportionately increase the operating cost of com- panies over a certain size. One of the textbook examples is France’s “social plan” rule that makes it compulsory for companies with at least 50 employees to make plans for the facilitation of reemployment. The social plan increases labour costs on the threshold, and it keeps many firms employing 30-40 people from going over 49 employees (see Figure 2 in Garicano, Lelarge & Van Reenen, 2013). To avoid such traps, certain EU poli- cies are targeted at high growth small and medium-sized enterprises (Wach, 2015).
The taxation system can be interpreted as an institutional factor, and as such it can influence the firm structure of the economy. It may favour certain types of business entities or certain firm sizes. This paper addresses some European cases where the tax regime may have a distortion effect on the size structure of businesses.
7.2. LITERATURE REVIEW
The literature on the effect of taxation on firm size is not vast. Henrekson & Johansson (1999) used firm data for Sweden for 1968-1993 to check how different size groups of firms evolved in the country. They have found that while the number of smaller firms (employing less than 10 people, and especially the ones employing 0-1 person) increased dynamically, the number of larger firms almost didn’t change at all, or even dropped in certain categories. They offer several explanations to the phenomenon (all of them being institutional ones), and one suggestion is specifically related to the Swedish tax system, which – according to the authors – disfavoured new, small firms, while large firms were highly favoured until the early 1990s. Since smaller firms could not grow because of the institutional environment, Sweden was left with a relatively low number of large firms, and a high number of slowly growing small ones (Henrekson & Johansson, 1999).
Heshmati, Johansson & Bjuggren (2009) also test Swedish firm size data for the peri- od 1973-2002, and they try to show the effect of corporate taxes on the size distribution
of firms. During that period Sweden had used various tax systems, the nominal corporate tax rate had varied between 62% and 28%. Using data acquired from balance sheets and income statements the authors were able to calculate the effective corporate tax rate for firms of various sizes. They found yet again that the tax system had favoured larger firms: the highest effective tax rate was paid by firms with 10-19 employees, while the lowest by ones with more than 2500 employees. The effective tax rate was also lower for firms in the production sector than for those operating in the service sector. These find- ings can be interpreted the same way as the results of the previous Swedish analysis: the taxation system prevented smaller firms to grow (as the highest corporate taxes were paid by the 10-19 employee group); Sweden was left with many microfirms and very few large ones.
Somewhat similar results were found by Chongvilaivan & Jinjarak (2010) in a cross- country analysis. They found that there is a positive empirical association between firm size and corporate tax, more precisely the authors state that: “raising the corporate tax rates by 5% increases the average growth of firm size by 3%” (Chongvilaivan & Jinjarak, 2010, p. 158). Once they split the sample into an OECD and a non-OECD group, however, the same effect was not found for the non-OECD countries. Since this paper focuses on the EU28, and since most EU countries are also members of the OECD, Chongvilaivan &
Jinjarak’s findings are relevant for our analysis.
One of the main issues in cross-country studies is the difference in the firm size dis- tribution between developed and developing countries. Many studies suggest that the difference can be caused at least partially by tax avoidance. Since the legal system is weaker in developing countries, the informal sector can be very widespread, constituting as much as half of the economic activity (Dabla-Norris, Gradstein & Inchauste, 2005), distortions in firm structure can lead to misleading results. In a study on Cote d’Ivoire Richmond and Klapper (2010) found that “the distorted firm size distribution relative to developed countries is systematic across all sectors of the economy” (p. 1). Given that the EU28 members are not homogeneous as far as the rule of law or the level of corrup- tion is concerned, similar distortions may occur in our analysis as well.
A relatively extensive literature exists on how tax incentives (especially ones target- ed at boosting R&D) affect firms of different sizes. In a Japanese study testing more than 900 companies for the period 1989-1998 it is found that large or larger firms can benefit more from such tax incentives (Koga, 2003). Labeaga Azcona, Martínez-Ros & Mohnen (2014) checked Spanish manufacturing firms, and while they had also found that large firms use the incentive more, the price elasticity in case of SMEs is larger than in case of large firms. In a final report compiled by the CPB Netherlands Bureau for Economic Policy Analysis (2014) it is remarked that: “In some of the countries analysed, small and medi- um sized enterprises (SMEs) tend to respond more strongly to the support for R&D, while the reverse was found in other countries” (p. 6). It seems that the effect of tax incentives on firm size is inconclusive in Europe.
7.3. MATERIAL AND METHODS
The paper uses several secondary sources, statistical databases to test the relationship between the size of firms and taxation:
− One of the main sources of the firm demographics data was Eurostat’s Structural Business Statistics database: http://ec.europa.eu/eurostat/web/structural-business- statistics;
− The second main source of EU country-level firm data was the 2014 issue of the European Commission’s SME Performance Review: http://ec.europa.eu/enterprise/
policies/sme/facts-figures-analysis/performance-review/index_en.htm;
− Macroeconomic data, such as GDP was collected from the IMF’s World Economic Outlook (IMF-WEO) database: http://www.imf.org/external/ns/cs.aspx?id=28;
− Macro-level taxation indicators were collected form the World Bank’s Doing Busi- ness database: http://www.doingbusiness.org/data/exploretopics/paying-taxes;
− For country-level tax information KPMG’s (http://www.kpmg.com/Global/en/ ser- vices/Tax/Pages/default.aspx) and PWC’s (http://www.pwc.com/gx/en/paying- taxes/) tax databases were used.
The materials used have their limitations. Most firm-related data is collected in groups based on the number of employees. As many taxation rules are based on the turnover of the firm, the number of employees is not the best indicator for the size of the firm in our analysis. Some information on turnover was also available, however the data was also organised according to the number of employees. The other problem with the groups formed according to the number of employees is that we only have a limited number of categories, namely firms with 0-9, 10-49 (10-19 and 20-49 in some cases), 50- 249 and 250+ employees.
The sources of firm size data were the Structural Business Statistics (SBS) and the SME Performance Review (PR) databases. Although PR takes most of its data from SBS, it also gathers some additional information from national statistics offices, and some esti- mates are also used, so the two databases are not completely compatible. It also has to be mentioned that not all of the businesses are counted in SBS and PR. The SBS database features active enterprises in the “business economy except activities of holding compa- nies”, while PR includes active enterprises from sectors B-J and L-N according to the revision 2 of NACE, i.e. it excludes firms operating in the agricultural, financial, education, health etc. sectors.
Basic data analysis methods, such as linear regression were used to test the relation- ships. The calculations and graphic interpretations were made with MS Excel.
Firms in the study are grouped into four categories. For the sake of simplicity firms employing 0-9 people are called micro firms; those employing 10-49, small firms; medi- um-sized are the ones with 50-249 employees; and large ones with at least 249 employ- ees.
7.4. RESULTS AND DISCUSSION
It is not surprising that the number of firms operating in an economy is correlated with the GDP of the country. But if we break down the total number of firms into different size groups (see Table 7.1.), we will find that the GDP is correlated a lot stronger to the number of larger firms than to those of the small ones. In case of the EU28 the Pearson’s R for firms with 0-9 employees and GDP correlation is 0.84, while it is 0.95 for larger firms (for small firms it is 0.97, for medium-sized ones it is 0.95 and for large ones it is
0.96, to be precise). The relationship is significant at all relevant significance levels (p<0.001).
Table 7.1. Number of firms and the GDP of the EU28 countries in 2015 Country
GDP (current int $ PPP,
billions)
Total no.
of firms
Firms with 0-9 em- ployees
Firms with 10-49 em-
ployees
Firms with 50-249 employees
Firms with 250+ em-
ployees
Austria 402.420 314700 275541 33143 4970 1049
Belgium 492.267 519513 488385 26299 4033 796
Bulgaria 131.334 308675 282432 21730 3945 570
Croatia 89.674 149968 137913 10014 1648 394
Cyprus 27.700 38992 35954 2547 416 76
Czech Rep. 325.285 1027192 988741 31130 5958 1361
Denmark 255.866 215749 193260 18752 3136 601
Estonia 36.845 60004 54362 4577 920 147
Finland 224.846 221392 203465 14794 2562 568
France 2633.896 2603509 2428690 146837 22828 5154
Germany 3815.462 2327524 1901634 354959 60067 10864
Greece 294.014 613971 591111 20148 2302 407
Hungary 255.254 521429 493595 23058 3958 817
Ireland 237.629 142949 126640 13498 2395 412
Italy 2157.123 3543835 3359266 162900 18692 2981
Latvia 49.731 94320 84006 8440 1680 192
Lithuania 82.622 142700 129771 10488 2170 268
Luxembourg 53.174 30300 26283 3248 623 149
Malta 14.716 32019 30342 1315 303 57
Netherlands 818.249 810214 759021 41376 8397 1424
Poland 996.477 1456927 1387965 51900 14280 2783
Portugal 287.388 757843 723760 29138 4261 687
Romania 406.964 442568 387827 46045 7334 1361
Slovakia 158.428 456389 439187 14621 2149 432
Slovenia 62.949 111914 105528 5214 980 189
Spain 1619.093 2229626 2107621 105349 14008 2647
Sweden 464.264 705430 667866 31056 5487 1020
United Kin. 2641.432 1766734 1568508 164210 28009 6009 Source: Own compilation based on IMF-WEO and PR data
The mentioned results might suggest that larger firms (with more than 10 employees) are more important for economic performance. A further investigation of the topic, however, can lead to even more convincing results. Figure 7.1. shows the distribution of firms in the EU28 countries. German micro firms (employing 0-9 people) have the lowest share of the total (around 82%), while Greek, Czech or Slovak micro firms have a share of 96%. Large firms (employing more than 250 people) make up only a fraction of the total number of enterprises. In Luxembourg, where large firms have the highest proportion in Europe, they have a share of 0.5%.
Figure 7.1. Firm size structure based on the number of people employed in the EU28 in 2015
Source: Own calculations based on PR data.
If the relationship between the GDP/capita and the firm size distribution is tested, the results clearly suggest that the higher the share of larger firms in an economy, the higher the GDP/capita is in that country. A negative correlation is found in case of micro firms (R= -0.39; p=0.04), while the correlation is positive and significant for all other groups of firms employing more than 9 people. The highest positive correlation is found when the share of large companies is tested, in case of which Pearson’s R is 0.58 (p=0.001).
0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
Greece Czech Republic Slovakia Portugal Poland Italy Malta Sweden Hungary Spain Slovenia Belgium Netherlands France EU28 Cyprus Croatia Finland Bulgaria Lithuania Estonia Denmark Latvia United Kingdom Ireland Romania Austria Luxembourg Germany
0 - 9 10 - 49 50 - 249 250 +
Firm Size and Corporate Taxation
Information included in the World Banks’ Doing Business database on taxes paid by the enterprises was used to check the interrelation between firm size and taxation. The da- tabase contains three categories that are relevant for our inquiry (see Table 7.2.):
− Tax rates, given as a percentage of the commercial profit, and broken down into three groups: profit tax, labour tax and contribution, and other taxes;
− Number of payments per year;
− Time requirement for paying taxes (the time it takes to prepare, file and pay the corporate income tax, value added or sales tax, and labour taxes, including payroll taxes and social contributions – in hours per year).
The data show that tax rates are negatively correlated with corporate size. The cor- relation between the share of micro companies and the total tax rate is positive, alt- hough not significant (p=0.10). For the share of all other size groups, though, there is a negative correlation, and it is significant, too, for medium-sized (R=-0.4; p=0.03) and large (R=-0.41; p=0.03) firms. Although the Pearson’s R value only suggest a moderate correlation, one has to conclude that higher taxes paid by firms tend to coincide with a higher proportion of micro firms in the economy, which, on the other hand, coincides with a lower GDP/capita value.
The phenomenon might partially be explained by tax avoidance (see the works of Dabla-Norris. Gradstein & Inchauste, 2005; and Richmond & Klapper, 2010 cited above).
Higher taxes can create an incentive to increase illegal employment, and that will shrink the official size of firms. Since the DB database breaks down the total taxes paid by com- panies into three groups, the two significant of which is profit tax and labour tax, we can test which one of these two is more closely correlated with the share of firms of different sizes. It is found that profit tax is not significantly correlated with firm size. Labour tax, on the other hand, has the same effect on firm size as the total tax had: R=-0.41 & p=0.03 for medium-sized firms; R=-0.43 & p=0.02 for large firms. Higher labour taxes negatively affect the share of medium-sized and large firms in an economy. The effect might be caused because of illegal employment, or it could simply be that with higher taxes on employment it doesn’t pay off to increase the number of employees after a certain limit.
Taxes can distort firm operation not only because of the rates, but also because of the transaction costs they generate for firms. The DB database has two indicators for those transaction costs: number of payments per year and number of hours it takes in a year to handle the chores related to tax payments. No significant association is found between the number of payments and the size distribution. In case of the time require- ments, however, the same relation is found as in case of the tax rates. More time spent on administering taxes means a higher share of micro firms, but the significance level is above .05 (R=0.35 & p=0.07). Negative Pearson’s Rs are found for small and medium- sized firms, with the significance value again around 0.07. But in case of large firms the correlation is negative and significant (R=-0.41 & p=0.03), showing that the more time has to be spent on administering tax payments, the less share large firms will have in the economy. The result is an indication that the lowering of transaction costs might help firms to grow bigger, which, on the other hand can increase the wealth generated by the economy (measured in GDP/capita).
Table 7.2. Paying corporate taxes in the EU28 countries in 2013
Country Payments
(nr. per year)
Time (hours per year)
Profit tax (%)
Labour tax (%)
Other taxes (%)
Total tax rate (%)
Austria 12.00 166.00 15.40 34.30 2.40 52.00
Belgium 11.00 160.00 6.50 50.70 0.60 57.80
Bulgaria 13.00 454.00 5.00 20.20 1.70 27.00
Croatia 19.00 208.00 0.00 17.10 1.60 18.80
Cyprus 29.00 146.50 9.60 12.00 1.60 23.20
Czech Republic 8.00 413.00 7.60 38.40 2.60 48.50
Denmark 10.00 130.00 20.30 3.00 2.80 26.00
Estonia 7.00 81.00 8.40 39.00 1.90 49.30
Finland 8.00 93.00 14.50 24.20 1.30 40.00
France 8.00 137.00 7.40 51.70 7.50 66.60
Germany 9.00 218.00 23.30 21.20 4.30 48.80
Greece 8.00 193.00 18.20 31.00 0.70 49.90
Hungary 11.00 277.00 11.80 34.30 1.80 48.00
Ireland 9.00 80.00 12.40 12.10 1.40 25.90
Italy 15.00 269.00 19.90 43.40 2.00 65.40
Latvia 7.00 193.00 4.90 27.20 2.90 35.00
Lithuania 11.00 175.00 6.10 35.20 1.30 42.60
Luxembourg 23.00 55.00 4.20 15.60 0.40 20.20
Malta 7.00 139.00 30.30 10.70 0.50 41.60
Netherlands 9.00 123.00 21.10 17.60 0.40 39.00
Poland 18.00 286.00 13.10 24.70 1.00 38.70
Portugal 8.00 275.00 15.10 26.80 0.50 42.40
Romania 14.00 159.00 10.70 31.50 1.00 43.20
Slovakia 20.00 207.00 8.50 39.70 0.40 48.60
Slovenia 11.00 260.00 12.50 18.20 1.40 32.00
Spain 8.00 167.00 21.90 35.70 0.70 58.20
Sweden 6.00 122.00 13.40 35.50 0.60 49.40
United Kingdom 8.00 110.00 20.90 11.30 1.50 33.70
Source: World Bank’s Doing Business Database
Using KMPG’s tax database we can make another attempt at measuring the com- plexity and hence the transaction costs imposed by the tax system. KPMG adds short comments to the corporate tax rate on its website to provide additional information. The comment in some cases is quite long; in case of Spain, for example it is 2,053 characters long (see Figure 7.2.). In case of other countries like Finland or Bulgaria KPMG does not feel the need to make additional comments. If we assume that longer comments are needed in case of more complex legislation, and we count the number of characters needed by KPMG to describe the corporate tax system, the length of the comments can be used as an indicator of corporate tax complexity.
Figure 7.2. Number of characters used by KPMG to describe the corporate tax system Source: Own calculations based on KMPG database.
When the association is tested between KPMG’s comment length and firm size dis- tribution data, one significant relationship is found. The PR database provides infor- mation on the turnover generated by firms of different sizes (according to the number of people employed). Using this information a similar distribution of micro, small, medium- sized and large firms can be set up, as it was done in case of employment figures. In the UK almost 56% of the total turnover was generated by large firms in 2011; in Germany it was 52%, while in Cyprus it was only 19.5% (Malta and Luxembourg were excluded from the analysis for lack of sufficient data). It was found that there is a negative association between corporate tax comment length and the share of medium-sized (50-249 employ- ees) firms’ turnover from the total turnover (R=-0.39 & p=0.045). The more complex corporate tax system, the lower share of turnover of medium-sized companies there is.
The numbers also suggest that the complexity does not have a significant influence on the turnover of either micro and small firms, or large ones.
0 500 1000 1500 2000 2500
Finland Bulgaria Slovakia Sweden Poland Netherlands Luxembourg Latvia Austria Slovenia Denmark Estonia Cyprus Germany Ireland Belgium Romania Greece Lithuania United Kingdom Malta Hungary Portugal Croatia Italy France Czech Republic Spain
Firm Growth and Taxation
The SBS database provides information on firm growth. It collects data for firms that have increased the number of their employees by at least 10% over the past calendar year. The latest available data is for the year 2012 (see Table 7.3.). The EU28 is reduced to EU27 here, because no data was available for Greece.
Table 7.3. Firms that increased the number of their employees by at least 10% in 2012 Country Average nr. of employees in
high growth firms
High growth firms/total number of firms ratio
Austria 70 0.8%
Belgium 67 0.5%
Bulgaria 86 0.9%
Croatia 103 0.3%
Cyprus 77 0.5%
Czech Republic 105 0.5%
Denmark 71 0.9%
Estonia 79 0.7%
Finland 63 0.9%
France 103 0.6%
Germany 83 1.5%
Hungary 90 0.6%
Ireland 82 1.1%
Italy 66 0.3%
Latvia 69 1.2%
Lithuania 78 1.4%
Luxembourg 54 1.4%
Malta 44 0.6%
Netherlands 82 0.8%
Poland 108 0.6%
Portugal 75 0.4%
Romania 141 0.3%
Slovakia 101 0.5%
Slovenia 70 0.5%
Spain 69 0.4%
Sweden 68 0.8%
United Kingdom 114 1.3%
Source: Own calculations based on the SBS database
With the help of the data provided by SBS, the effect of taxation characteristics on firm growth can be tested. Correlations were checked, but very few significant relations were found. There is a negative correlation between KPMG’s comment length and the ratio of high growth firms (R=-0.37 & p=0.06). The significance level is above 0.05, yet this is the most consistent relationship that could be established in case of the firm growth data. The correlation is less than moderate, and it might suggest that more com- plex tax systems limit firm growth, as the ratio of firms that increase their employment by more than 10% year on year is lower.
The correlation between the labour tax rate and the high growth firm ratio is also negative (R=-0.33 & p=0.1). The profit tax rate is also negatively correlated with the av- erage number of employees in high growth firms (R=-0.27 & p=0.17). Since the signifi- cance value is way over 0.05 in the latter two cases, these findings are not worth discuss- ing any further.
7.5. CONCLUSIONS
At the beginning of the analysis it was found that while the share of micro firms (0-9 employees) is negatively correlated with the GDP/capita of the EU28 countries, the share of larger firms, and especially that one of the large firms (employing more than 249 peo- ple) is positively correlated with it. Larger firms are clearly critical for economic perfor- mance. The paper tested whether taxation characteristics influence the size distribution of firms in the EU28 countries. Our findings may be concluded in three points:
− Firm size and corporate tax rates are negatively correlated in case of medium-sized (R=-0.4; p=0.03) and large (R=-0.41; p=0.03) firms as well. The higher the corporate tax rate, the less medium-sized and large companies there are in the economy. It was also found that from the taxes that have to be paid by companies, it is the la- bour tax and contribution that causes the negative correlation. One of the possible explanations can be that higher labour taxes create the incentive for firms to in- crease illegal employment, distorting the distribution of firms calculated from official statistics. Another explanation could be that as a result of higher taxes firms simply grow slower. This was also tested later on.
− Indications were found that higher transaction costs caused by taxation lead to smaller firms. Transaction costs can be measured as the number of hours required per year to administer tax payments, or as the length of additional comments need- ed in KPMG’s website to describe the corporate tax rate. A significant negative cor- relation is found between the number of hours related to tax payment, and the share of large firms (R=-0.41 & p=0.03), and also between KPMG’s comment length and the share of medium-sized firms’ turnover from the total turnover (R=-0.39 &
p=0.045). Lowering the transaction cost of corporate taxation might increase the firm size, and economic performance through that.
− More complicated tax rules might cause a smaller proportion of firms to grow quick- er. A negative correlation was found between KPMG’s comment length and the ratio of high growth firms (R=-0.37 & p=0.06). Since the significance level of all correla- tions involving firm growth are above p=0.05, no real connection could be estab- lished between firm growth and taxation from the data used.
There seems to be a contradiction between the findings of this paper and the ones of some of the other papers mentioned in the literature review. Chongvilaivan & Jinjarak (2010) argue, for example that there is a positive empirical association between firm size and corporate tax. Yet this paper states that the firm size and corporate tax rates are negatively correlated in case of medium-sized and large firms. The contradiction proba- bly comes from the way we measure large or small firms. This paper measures large firms with a ratio: number of large firms in a country divided by the total number of firms in the country. All firms that employ at least 250 people are counted as large ones.
Chongvilaivan & Jinjarak on the other hand use the OSIRIS database as a source, so they work with the data of 40,863 individual companies. They don’t have to work with groups of companies, and so a firm with 250 and a firm with 25,000 employees is obviously not counted as being similar and basically the same.
There is a different rationale behind the arguments as well. Chongvilaivan & Jinjarak assume that higher taxes lower the number of perfectly competitive firms, and that on the other hand raises the size of those few firms. The chosen way of measurement is well suited to test that argument. But there are other ways to look at the problem. One is to say that higher taxes slow down the growth of newly established firms. Another is to say that higher taxes create incentives for illegal employment. When groups of com- panies, and their share from the total is chosen as a measurement method, these argu- ments are easier to test and to back.
It also has to be mentioned that dynamic relationships were not tested in the analy- sis. The relationships checked were between taxation data characterizing the current state of an economy, and firm size structure that represents the current situation yet again. It is obvious that the firm distribution will not reflect immediately the tax rules; so if there is a change in the tax system, it will take a certain number of years to the firm structure to change. Dynamic relationships of such character could only be tested if cross-period analysis is used.
REFERENCES
Chongvilaivan, A., & Jinjarak, Y. (2010). Firm Size and Taxes. The Journal of the Korean Economy.
11(1): 145-175.
CPB Netherlands Bureau for Economic Policy Analysis (2014). A Study on R&D Tax Incentives. TAX- UD/2013/DE/315. FWC No. TAXUD/2010/CC/104 Final Report.
Dabla-Norris, E., Gradstein, M., & Inchauste, G. (2005). What Causes Firms to Hide Output? The Determinants of Informality. IMF Working Paper nr. 05/160.
Garicano, L., Lelarge, C., & Van Reenen, J. (2013). Firm Size Distortions and the Productivity Distri- bution: Evidence from France. NBER Working Paper 18841.
http://www.nber.org/papers/w18841
Gubik, A.S., & Bartha, Z. (2014). SME Internalisation Index (SMINI) Based on the Sample of the Visegrad Countries (chapter 2). In: Gubik. A.S. & Wach. K. (eds). International Entrepreneur- ship and Corporate Growth in Visegrad Countries. Miskolc: University of Miskolc. pp. 23-40.
Gubik, A.S. (2001). A Comparison of Experts' and Entrepreneurs' Opinions on International Busi- ness Activity. Theory, methodology, Practice, 7(2): 47-54.
Gubik, A.S. (2007). A kis- és középvállalatok nemzetközi megjelenésének esélyei: egy empirikus felmérés tapasztalatai. In: Gubik Andrea (ed.): VI. Nemzetközi Konferencia a közgazdász képzés megkezdésének 20. évfordulója alkalmából. pp. 312-321.
Gubik, A.S. (2008). Small and Medium Sized Enterprises Going International: The Case of Borsod- Abaúj-Zemplén County. In: Borowiecki R, Jaki, A. (eds.): Enterprises in the Face of 21st Century Challenges: Development – Management – Entrepreneurship. pp. 355-368.
Henrekson, M., & Johansson, D. (1999). Institutional Effects on the Evolution of the Size Distribu- tion of Firms. Small Business Economics. 12(1): 11-23.
Heshmati, A., Johansson, D., & Bjuggren, C.M. (2009). Effective Corporate Tax Rates and the Size Distribution of Firms. TEMEP Discussion Paper No. 2009:23.
Koga, T. (2003). Firm size and R&D tax incentives. Technovation. 23(7): 643-648.
Labeaga Azcona, J., Martínez-Ros, E., & Mohnen, P. (2014). Tax incentives and firm size: effects on private R&D investment in Spain. MERIT Working Papers no. 081.
Lee, J. (2009). Does Size Matter in Firm Performance? Evidence from US Public Firms. International Journal of the Economics of Business. 16(2): 189-203.
Pervan, M., & Visic, J (2012). Influence of Firm Size on Its Business Success. Croatian Operational Research Review. 3: 213-223.
Richmond, C., & Klapper, L. (2010). The Political Economy of Firm Size. CSAE Conference 2010.
Economic Development in Africa.
Symeonidis, G. (1996). Innovation. Firm Size and Market Structure. OECD Economics Department Working Papers No. 161. http://dx.doi.org/10.1787/603802238336.
Urbaniec, M. (2015). Europeanization of Entrepreneurship and SME Policy: Challenges and Oppor- tunities (chapter 5). In: P. Stanek & K. Wach (Eds.). Europeanization Processes from the Mesoeconomic Perspective: Industries and Policies. Kraków: Cracow University of Economics, pp. 95-111.
Vaona, A. & Pianta, M. (2006). Firm size and innovation in European manufacturing. Kiel Working Paper No. 1284.
Wach, K. (2015). Modern Policy for the Entrepreneurial Economy: Theoretical Considerations (chapter 1). In: A.S. Gubik & K. Wach (Eds.). Institutional Aspects of Entrepreneurship. Miskolc:
University of Miskolc, pp. 9-18.
ONLINE DATABASES AND SOURCES
Eurostat’s Structural Business Statistics database (SBS): http://ec.europa.eu/eurostat/web/ struc- tural-business-statistics
European Commission’s SME Performance Review (PR, 2014): http://ec.europa.eu/enterprise/
policies/sme/facts-figures-analysis/performance-review/index_en.htm
IMF’s World Economic Outlook (IMF-WEO) database: http://www.imf.org/external/ns/cs.aspx?id
=28
World Bank’s Doing Business database (DB): http://www.doingbusiness.org/data/exploretopics/
paying-taxes
KPMG database: http://www.kpmg.com/Global/en/services/Tax/Pages/default.aspx PWC database: http://www.pwc.com/gx/en/paying-taxes/
ACKNOWLEDGEMENTS AND FINANCIAL DISCLOSURE
The chapter was prepared within the research project OTKA K 109839, entitled ‘Entrepreneurial intentions of Hungarian students and possible incentives’ (2013-2016)
